> The current work in progress is available at :
> http://people.freebsd.org/~murray/
> Any feedback would be greatly appreciated.
>
> Thanks,
>
> - Murray
Okay, I read your page and printed it out, and went over it a few times. A
couple of things bothered me, but for the most part I agreed with it. Then
I showed it to a few friends of mine and they seemed to be bothered by the
same statements so I finally decided I would bring them to your attention.
In order for your document to be considered a reasonable and unbiased paper,
it must be reasonable and unbiased :) Some of your statements leave us with
the taste of the Anti-Microsoft-Jihad. For example:
> All Windows users are familiar with the "Blue Screen of Death". Poor
reliability is one of the major drawbacks of Windows. Some of the major
issues have been fixed in Windows 2000, but "code bloat" has introduced many
more reliability problems.
I run Windows 2000 on many desktop systems, and I've run Windows 2000 Server
and Advanced Server. So have many friends of mine. I have seen FEW BSODs
in Windows 2000. About as many BSODs as I've seen FreeBSD kernel panics or
hardware driver lockups. You can't say this statement like this, implying
that Windows 2000 BSODs, and then say some of the major issues have been
fixed in 2000. Windows 2000 isn't a "fix" of Windows 98. It's basically
Windows NT 5.0. I know Windows NT 4.0 Workstation to be a very rock stable
OS for basic workstations. Then you introduce the term "code bloat", which
is a favorite in Anti-Microsoft rants. "Code bloat" apparently means
implementing more code/features/etc than a certain computer "expert" feels
should have been included. It also implies that this extra "bloat" causes
instability or reduced speed in the product. Certainly, some operating
systems, like Windows Millennium, could be described as "bloated", however,
to imply that Windows 2000 has some sort of "code bloat" when it actually
outperforms the same software applications under Windows 98 is completly
nonsensical. Yes, Windows 2000 does take a more powerful computer to run
than Windows 98. But that is comparing apples to oranges. How much more
powerful of a computer does Windows 2000 take than Windows NT 4.0 ? When
considered in this light, Windows 2000 (NT 5) doesn't require much more of a
hardware upgrade than many other operating system upgrades. Consider the
fact that Windows NT 4.0 to Windows 2000 (5.0) is certainly no small update,
one would expect the OS, being much more powerful, to require more
resources. So, consider revising this statement. I've found that when you
treat it properly, Windows 2000 can be a very stable operating system.
Unintelligent administrators can certainly create an unstable Windows 2000
setup, but that doesn't nessisarily detract from the OS itself. If you're
going to be critical of Windows 2000's reliability, you need to support it
better than you currently are. You're also very quick to talk about
FreeBSD's Softupdates, but you fail to say anything reasonable about Windows
2000's disk access speed or reliability. If you slant your judgement so far
against the other products, it makes you sound like you don't know what
you're talking about (no offense). You need to point out the pros and cons
of ALL three systems. Not just the pros of FreeBSD and the cons of
Linux/Windows. NO ONE in their right mind believes that FreeBSD is the
beginning and the end, the one and only true operating system, master of all
features, speed, and reliability. It's just our favorite and it has certain
major advantages. That doesn't mean you can ignore the advantages of the
other operating systems.
> Windows is adequate for routine desktop apps, but it is unable to
handle heavy network loads.
A broad statement with no support. Define heavy network loads.
> A few organizations try to make it work as an Internet server.
Only a few ?
> For instance, barnesandnoble.com uses Windows-NT
Now you're going to cite Windows NT 4 in your argument against Windows 2000
? Your comparison is FreeBSD vs Linux vs Windows 2000, not Windows NT 4.0.
That would be like me saying, blahblahblah.com is using Red Hat Linux 5.0
with Wu-FTPd 2.6 so Linux is totally insecure.
> For their own "Hotmail" Internet servers, Microsoft used FreeBSD for
many years.
Although this is somewhat true, how long has Microsoft owned Hotmail ?
Didn't they plan to remove all FreeBSD servers and replace them with Win2k
servers from the beginning ? Isn't that a massive project to move a live
service like that ? Haven't they completed that project and removed all
FreeBSD servers (requiring more Win2k servers than former BSD servers to
handle the load of course) ? Once again, you made a broad statement that
implies a lot, but you are somewhat misleading the reader by not giving
enough fact. If you don't have enough room to completely explain a
complicated statement like this, I suggest you not include it.
> The Microsoft FAT filesystem and the newer NTFS are both plagued by
over 15 years of backwards compatability with the earliest of PC-based
filesystems.
How is NTFS plagued with over 15 years of backwards compatability ? I
wasn't aware of NTFS being backwards compatible with anything.
> These filesystems were not designed for today's demanding server
applications, they weren't even designed with a multi-user OS or networking
in mind!
I believe NTFS _was_ designed with multi-user operating systems and
networking in mind. Whether they succeeded or not is another question, but
I have yet to see a benchmark between Windows 2000 NTFS, Windows 98SE FAT32,
Linux ext2, and FreeBSD FFS. You can't slam something like NTFS with an
untrue statement and no support.
> There are often conflicts when using a device driver on different
versions of Microsoft Windows
This statement has no value, because it's obvious. It would be like trying
to use a KLD from one version of BSD with another. Of course it's probably
not going to work. You can't state this like it's a "con" of using Windows.
> The amount of free Windows software is much less than what is available
for Unix.
I almost choked to death on my Submarina Sandwich when I read this. I think
you need to take a step back and think a bit on this one. Do you really
think the 4000-5000 ports in the ports tree and the other misc free
applications on the net outnumber the hundreds of thousands of free Windows
applications ? Even if you don't count Shareware, which you really should,
the number of free Windows applications outnumbers the number of free Unix
applications 10 to 1. Now, the power, importance, usefulness, etc of the
Unix apps may be superior, but the "amount" or number of free Windows
applications is certainly not "much less" than what is available for Unix.
It's much much more. Giving Windows a :( for Free Applications is
absolutely unbelievable. Oh, and if you're going to include all of the
Linux binary compatible applications as "free software for FreeBSD", then
you would have to include in Windows 2000's count all of the DOS binaries it
is compatible with, which far far outnumbers any amount of software written
for any operating system ever. The amount of freeware produced between
MSDOS 3 and MSDOS 6 is uncountable, and much of it is compatible with
Windows 2000.
> Very few development tools are included with Windows 2000. Most need to be
purchased separately, and are rarely compatible with each other.
Another :( for Microsoft in the area of "Development environment". While
your statement is certainly true, you forget to mention that the development
ENVIRONMENT of the Microsoft compatible development tools is actually far
superior. I love using Microsoft Visual C++ 6.0 and even Borland Builder
sometimes. Show me anything on the level of Microsoft Visual C++ 6.0 for
Unix. Some people may like to write their C code in VI or VIM and compile
it with GCC using a make file, but that certainly doesn't make it the nicest
"development environment". And what's this about the development tools
rarely being compatible with each other ? Insane. I can make a DLL in
Borland Builder and call it from a program written in Visual C++, and I can
import Win32 API code from Visual C++ to Borland Builder or any other Win32
API compiler. This statement you've made, and the :( are terribly unjust
and biased against the Microsoft Windows development tools. I would like to
see something on the level of Borland Turbo C++ 3.0 for DOS, in the way of a
Unix IDE before you start talking about poor development tools. The
compiler may be packaged seperately, but that is more of an economics issue.
You make it sound like it's terribly difficult to code in windows because of
poorly written and incompatible tools. You also fail to mention that
Microsoft's compiler's optimizations don't produce unstable code like GCC,
and that Microsoft's compiler can plug in compilers from CPU manufacturers
which generate code optimized for certain CPU extentions, like MMX, 3d-Now!
and SSE2.
> Very little is known about the internal development infrastructure of
Microsoft but the "blue-screen of death" speaks for itself.
Another unsupported BSOD generalization.
> Although support is available for Windows 2000, you should be prepared to
spend as long as an hour on hold, with no guarantee that your problem will
be resolved.
And you fail to mention there are FAR more Windows support helplines than
FreeBSD or Linux support lines.
> Because of the closed source nature of Windows, there is no informal,
free support available
Insanely untrue. There are hundreds of webpages, mailing lists, chatrooms,
free magazines, and more dedicated to Windows support, optimization,
tweaking, frequent questions, etc. Far more than there are for FreeBSD I
daresay.
> Since Windows 2000 is not updated frequently, you may wait years for
bugs to be fixed.
I guess you've never visited http://windowsupdate.microsoft.com
They have "Critical Updates" posted every couple of DAYS, and of course,
they download and install automaticly. You must have missed the service
pack 1 update not too long ago.
> Documentation is expensive, and very little on-line documentation is
provided.
Windows 2000 documentation is expensive ? Very little on-line documentation
is provided ?
You need to check out:
http://support.microsoft.com/directory/
The Microsoft Knowledge Base is FAR larger than the FreeBSD handbook. Plus,
I wouldn't wager the number of FreeBSD manpages vs the number of Windows
Helpfiles. The Windows Helpfile format is often easier to search and index,
and allows hyperlinking to other pages quickly.
> System Startup :( Windows sucks ass in this respect.
I've gone back and checked out your page, and you have written this, which
is no doubt unfinished, but why exactly does Windows 2000 "suck ass" in
system startup ? What's wrong with the boot.ini setup ? You can use it to
boot MSDOS or Windows 98 or FreeBSD or Linux. Nothing "sucks ass" about it.
This is what troubles me about your document. You are intending to prove
that Windows "sucks ass", that Linux sucks just a little bit, and FreeBSD is
the almighty bullgod of operating systems. You haven't even defended this
statement, most likely because you have limited knowledge of the Windows
2000 bootloader, and yet your filler for this space is "sucks ass". You
already have the frown image in place. If you're going to make a document
like this, and expect people to take you seriously, you can't spread
misinformation and bias. That is the way of the Jihad idiots of other
operating systems. We don't want rabid FreeBSD kiddies. We want informed
unbiased FreeBSD users who know why FreeBSD is good, and where it lacks. As
it stands your document contains many many misleading statements, partial
truths, slightly untrue statements, and some outright falsehoods. I like
the idea of your document, and I'm interested in seeing it develop,
otherwise I wouldn't have spent the time writing you about it. My friends
and I all think it has the potential to be an excellent document, but if we
were to take it in front of knowledgable Linux or Windows 2000
administrators as it is, we would be slaughtered by the misstatements. Take
the time to read your document again, and try to remove any major
generalizations like the BSOD, and any statements detracting from Microsoft
without any factual support. You seem to be giving Linux and Solaris better
marks than Windows in areas where Linux and Solaris suck far worse than
Windows. Your paper is very Unix centric, and beyond that, very FreeBSD
centric. You have to be honest and let FreeBSD's power speak for itself.
Expand on the reasons why FreeBSD is powerful, rather than expanding on the
"reasons" why Windows 2000 "sucks ass".
Here's my flame retardant disclamer: As I said, if I wasn't interested in
seeing your document progress, I wouldn't have replied to you. I certainly
wouldn't spend all this time writing this message to try to flame you. I'm
trying to help you see that your document comes across as "rabid" when you
short Windows 2000's advantages, and YES it does have some advantages.
Having advantages is in no way admitting that it is a better OS. In other
words, the "best" OS doesn't have to be the best at EVERYTHING. If one
believes their OS is the best at EVERYTHING, one probably doesn't know
enough about operating systems in general to write any sort of document on
the subject. In any case, please take my email as contructive criticism,
rather than an attempt to attack or flame. Thanks.
Feel free to mail me back, public on the list, or privately on any of these
points. I'd love to help you draw up more fair lines to put in the boxes.
I hope some of the others on the list will put their two cents in too,
although lets try to keep this flame free ok ?
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-hackers" in the body of the message
